Business intelligence & analytics Customer intelligence Retail / loyalty American Express / Payback · LATAM · 2015–2020

Customer Loyalty Analytics Platform

The Payback loyalty program had millions of members – but no single view of any of them. Data was split across Santander, Banamex, Canadian Tire, Disney, Cinemex, and Interjet. Marketing teams were making retention decisions without data, and at-risk members were leaving before anyone noticed the pattern.

The challenge
The Challenge
Fragmented customer data across partner networks

The Payback loyalty program operated across dozens of retail and financial partners, each with separate data systems and formats. Customer behaviour data was siloed, making it impossible to build a unified view of member activity. Marketing teams were making decisions based on incomplete data, leading to ineffective retention campaigns and declining loyalty metrics in key segments.

The solution
The Solution
Centralised analytics layer with a 360° member view

Led the architecture and implementation of a centralised analytics data layer that ingested, normalised, and unified member data from all partner sources. Built ETL pipelines to consolidate transactional, redemption, and behavioural data. Designed a dimensional data model enabling recency-frequency-monetary segmentation, cohort analysis, and campaign attribution. Delivered an executive dashboard suite covering acquisition, engagement, and churn metrics across all partner categories.

The results
The Results
15% loyalty increase and faster campaigns

The unified analytics platform drove a measured 15% increase in customer loyalty metrics across the program. Marketing teams could now act on behaviour signals in near-real-time, cutting campaign design cycles from weeks to days. The 360° member view enabled personalisation at scale, reducing churn in at-risk segments by identifying behavioural patterns weeks before members would have lapsed.

15%
Customer loyalty increase across the program
Faster campaign design-to-launch cycles
360°
Unified customer view across the partner network
12+
Partner data sources integrated into a single layer
Technologies used
Python Apache Spark AWS Kinesis AWS S3 / Lambda Hadoop / Hive SAS Enterprise Guide SQL (complex) MongoDB Talend Tableau Power BI